Shou Guang Talent Recruitment Website Cover In The Retrieval Algorithm

The rapid development of computer, communication and network technology is affecting people's work and life. Facing massive data in the WEB, how to get the most valuable information is becoming the urgent matter. The data mining technology is a data processing technology that conforms to the need. Clustering is an important direction of research in the data mining. Clustering analysis in e-commerce, image processing and pattern recognition, text classification and other fields have a wide range of applications.Based on the research of the basic concept, process and the algorithm in data mining technology, and the current network resources in the search for sorting algorithm, I found many problems. Although the data mining algorithm has been used to in the process of network retrieval, it is widely used in the search engine of web now. But electronic data pretreatment, resume searching and the relevant literature in the recruitment of the network is less.In order to improve the efficiency of the resources recruitment, this paper puts forward a clustering algorithm that based on the effective index and dissimilarity. First, gain human resource data from the resume in network. Select the partial attributes of talent according to the needs about a clustering algorithm that based on the effective index and different index. And do data scrubbing, data pretreatment and data exchange etc. Second, calculate the different index of each object. And gain effective clustering initial object from the k sets of data in the biggest. That reduces number of iteration and improves the efficiency of clustering. Third, based on the different clustering and the combination of the clustering efficiently index that Ray proposed, adjust clustering parameter dynamically, and improve clustering validity and rationality. Fourth, the use of the algorithm in the categories from resources by clustering analysis is divided into small resources. And save it in sequence (based on the visit weekly) by counting the number of clicking. That influences sequence. Finally record realize to quick reasonable and effective human resource in shorter time.That provide technical support and more reasonable, humanized service for the enterprise personnel decisions.It is different from keyword search and fuzzy retrieval in the original site. This paper applies the data mining in the human resources advantage to personalized service in the web. Comparing with the original method, we can obtain inherent relation and implicit information of data attributes in human resources recruitment resume, In order to improve retrieval accuracy.